Triple

T3357818
Position Surface form Disambiguated ID Type / Status
Subject Mary Tyler Moore E70646 entity
Predicate spouse P13 FINISHED
Object Robert Levine
Robert Levine is an American cardiologist best known as the husband of actress Mary Tyler Moore.
 E354285 NE FINISHED

How this triple was built (4 steps)

Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.

NER Named-entity recognition gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: Robert Levine | Statement: [Mary Tyler Moore, spouse, Robert Levine]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07 
Target entity: Robert Levine
Context triple: [Mary Tyler Moore, spouse, Robert Levine]
  • A. Joseph Markovitch
    Joseph Markovitch was the father of French photographer and painter Dora Maar, a key figure in the Surrealist movement and companion of Pablo Picasso.
  • B. Stephen Frank
    Stephen Frank was an early American pioneer after whom the city of Frankfort, Kentucky, is named.
  • C. Ronald N. Perlman
    Ronald N. Perlman is an American actor best known for his distinctive deep voice and leading roles in works such as the "Hellboy" films and the television series "Beauty and the Beast."
  • D. Sam Denoff
    Sam Denoff was an American television writer and producer best known for his influential work on classic sitcoms such as "The Dick Van Dyke Show" and "That Girl."
  • E. Douglas Shulman
    Douglas Shulman is an American public official who served as the head of the U.S. Internal Revenue Service (IRS) during the late 2000s and early 2010s.
  • F. None of above. chosen
  • G. Unsure - the case is ambiguous/there is not enough information to decide.
NEDg Description generation gpt-5.1
Instruction
Generate a one-sentence description of the target entity. 
You are given a context triple in the form (subject, predicate, object), where the object is the target entity. 
# Instructions
Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. 
Avoid repeating the information from the triple, unless really essential.
# Response Format
Return only the sentence: "Description: [one-sentence description of the target entity]"
Input
Entity: Robert Levine
Triple: [Mary Tyler Moore, spouse, Robert Levine]
Generated description
Robert Levine is an American cardiologist best known as the husband of actress Mary Tyler Moore.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07 
Target entity: Robert Levine
Target entity description: Robert Levine is an American cardiologist best known as the husband of actress Mary Tyler Moore.
  • A. Joseph Markovitch
    Joseph Markovitch was the father of French photographer and painter Dora Maar, a key figure in the Surrealist movement and companion of Pablo Picasso.
  • B. Stephen Frank
    Stephen Frank was an early American pioneer after whom the city of Frankfort, Kentucky, is named.
  • C. Ronald N. Perlman
    Ronald N. Perlman is an American actor best known for his distinctive deep voice and leading roles in works such as the "Hellboy" films and the television series "Beauty and the Beast."
  • D. Sam Denoff
    Sam Denoff was an American television writer and producer best known for his influential work on classic sitcoms such as "The Dick Van Dyke Show" and "That Girl."
  • E. Douglas Shulman
    Douglas Shulman is an American public official who served as the head of the U.S. Internal Revenue Service (IRS) during the late 2000s and early 2010s.
  • F. None of above. chosen

Provenance (5 batches)

The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.

Step Stage Batch ID Status When
creating Elicitation batch_69ad85a660c48190998489309a3b4869 completed March 8, 2026, 2:20 p.m.
NER Named-entity recognition batch_69adb244435c81908e35d2aa36ec4f46 completed March 8, 2026, 5:30 p.m.
NED1 Entity disambiguation (via context triple) batch_69b34bbbc3a48190bc17572ba3136948 completed March 12, 2026, 11:26 p.m.
NEDg Description generation batch_69b34e45a6c08190a0011eaa60f3d50a completed March 12, 2026, 11:37 p.m.
NED2 Entity disambiguation (via description) batch_69b34eba517881908806b1ac285448ff completed March 12, 2026, 11:39 p.m.
Created at: March 8, 2026, 3:13 p.m.